Annexin A2 regulates TRPA1-dependent nociception

Abstract

The transient receptor potential A1 (TRPA1) channel is essential for vertebrate pain. Even though TRPA1 activation by ligands has been studied extensively, the molecular machinery regulating TRPA1 is only poorly understood. Using an unbiased proteomics-based approach we uncovered the physical association of Annexin A2 (AnxA2) with native TRPA1 in mouse sensory neurons. AnxA2 is enriched in a subpopulation of sensory neurons and coexpressed with TRPA1. Furthermore, we observe an increase of TRPA1 membrane levels in cultured sensory neurons from AnxA2-deficient mice. This is reflected by our calcium imaging experiments revealing higher responsiveness upon TRPA1 activation in AnxA2-deficient neurons. In vivo these findings are associated with enhanced nocifensive behaviors specifically in TRPA1-dependent paradigms of acute and inflammatory pain, while heat and mechanical sensitivity as well as TRPV1-mediated pain are preserved in AnxA2-deficient mice. Our results support a model whereby AnxA2 limits the availability of TRPA1 channels to regulate nociceptive signaling in vertebrates.

Keywords: TRPA1 channels; membrane abundance; nociception; protein–protein interaction.

Figure 1.

AnxA2 physically binds to TRPA1. A, AnxA2 coimmunoprecipitates with native TRPA1 from mouse sensory neurons as identified by MS/MS. The table depicts the MS/MS results (identified peptides and number of total spectra) of two independent TRPA1-affinity purifications and corresponding controls. B–D, Representative Western blots (WB) of immunoprecipitation (IP) experiments in HEK293T cells recombinantly expressing the indicated constructs. B, AnxA2 is detected in eluates (E) of immunoprecipitations only upon cotransfection of TRPA1-myc and AnxA2 (TRPA1-myc + AnxA2) but not in control conditions (TRPA1-myc + Mock). A deletion construct of AnxA2 lacking the first 15 aa (ΔAnxA2) did not coimmunoprecipitate with TRPA1 (TRPA1-myc + ΔAnxA2). Immunoprecipitations were performed with myc antibodies. C, P11 did not coimmunoprecipitate with TRPA1 in our assay (TRPA1-myc + p11). Immunoprecipitations were performed with myc antibodies. D, AnxA2 did not coimmunoprecipitate with TRPV1 in cotransfected HEK293T cells (TRPV1 + AnxA2) while TRPV1 itself is readily immunoprecipitated (TRPV1 is represented by the double band). Immunoprecipitations were performed with TRPV1 antibodies. I, input. Western blots were probed as indicated.

Figure 2.

AnxA2 neither affects TRPA1 voltage dependence nor cellular responses to the TRPA1 agonist MO. A, B, AnxA2 does not affect the voltage dependence of TRPA1. A, Whole-cell currents in response to voltage steps applied to HEK293T cells expressing TRPA1 (TRPA1 + Mock) or TRPA1 + AnxA2. B, Average ± SEM voltage dependence of TRPA1 peak tail currents at −75 mV for indicated transfections. For each transfection condition data were separately normalized to the current obtained after the maximum depolarization level (+175 mV). C, Representative currents elicited by I/V ramps after (MO) application of 25 μm MO for indicated transfections. D, Left, representative MO-induced current at −70 mV (holding potential). Right, Average ± SEM time constants of MO-induced activation and inactivation measured by a mono-exponential fit to the currents obtained at −70 mV for indicated transfections (n >12 cells per condition; n.s.). E, AnxA2 does not affect cellular responses to the TRPA1 agonist MO. Representative images of ratiometric [Ca⁺²]_i measurements in HEK293T cells expressing TRPA1 (TRPA1 + Mock) or TRPA1 and AnxA2 (TRPA1 + AnxA2). GFP was cotransfected to visualize transfected cells for further analysis. The image shows the cellular response to 10 μm MO (MO). One hundred micromolar ATP was applied after MO to control for cellular health. Scale bar, 10 μm. F, Dose dependency of MO-evoked increase of cellular [Ca⁺²]_i. All data are represented as mean ± SEM.

Figure 3.

AnxA2 is coexpressed with TRPA1 in nociceptors. A, B, Representative images of immunohistochemistry on cryosections of mouse DRG colabeled for AnxA2 and Peripherin in WT mice (A) and AnxA2^−/− (B) littermates. Scale bar, 20 μm. C–E, Representative images of immunohistochemistry on cryosections of rat DRG colabeled for AnxA2 and Peripherin, NF200, or TRPA1, respectively. White arrows indicate examples of neurons coexpressing AnxA2 and TRPA1. Scale bar, 40 μm.

Figure 4.

AnxA2^−/− mice exhibit more TRPA1-positive DRG neurons. A, B, Representative images (A) and quantification (B) of immunohistochemistry on DRG cryosections from AnxA2^−/− mice and WT littermates labeled for TRPA1, Peripherin, NF200, and TRPV1 as indicated. Scale bar, 20 μm. All data are represented as mean ± SEM.

Figure 5.

TRPA1 responses are sensitized in a subset of AnxA2^−/− sensory neurons. A–F, AnxA2^−/− DRG cultures are more sensitive to low MO concentrations as measured by ratiometric calcium imaging. A, The graph depicts the percentage of responders to 12 μm, 25 μm, or 50 μm MO (12 μm, AnxA2^−/−: 21.5 ± 1.5 compared with WT: 11.4 ± 1.7, p = 0.0008, Student's t test; n ≥ 500 neurons for each MO concentration from N = 3 independent cultures each). B, Representative averaged traces from all neurons in one coverslip (including responders and nonresponders to applied stimuli) upon application of indicated stimuli. C, Quantification of the percentage of neurons responding to each stimulus (MO; 12 μm, see data in A; 50 μm, AnxA2^−/−: 34.5 ± 1.9 compared with WT: 32 ± 2.7, n.s.; Cap 0.5 μm, AnxA2^−/−: 22.6 ± 2.8 compared with WT: 19.9 ± 2.9, n.s.; n = 560 WT and 640 AnxA2^−/− neurons, N = 3 independent cultures each). D, Quantification of response amplitudes to each stimulus (measured as peak increase over baseline). E, F, Quantification of the percentage of neurons responding to each stimulus (E) and the response amplitudes (F) in Mock-transfected WT neurons (WT Mock), Mock-transfected AnxA2^−/− neurons (AnxA2^−/− Mock), and AnxA2^−/− neurons transfected with AnxA2 cDNA (AnxA2^−/− rescue). Twelve micromolar MO (WT Mock: 17.1 ± 2.2, AnxA2^−/− Mock: 25.9 ± 3.4, AnxA2^−/− rescue: 12.4 ± 2.8; *p < 0.05, **p < 0.01, ANOVA with Newman–Keuls test; n = 292 WT Mock, 275 AnxA2^−/− Mock, and 269 AnxA2^−/− rescue neurons, N = 3 independent cultures each). Cap, capsaicin.

Figure 6.

AnxA2 restricts TRPA1 membrane levels in cultured DRG neurons. A, B, DRG cultures from AnxA2^−/− mice and WT littermates were nucleofected with mTRPA1 (top) or conucleofected with AnxA2 cDNA (AnxA2^−/− rescue; control: AnxA2^−/− Mock; bottom) and subjected to live labeling to selectively visualize TRPA1 channels at the plasma membrane. Representative images (A, taken at the bottom of the coverslip) and quantification (B) of live labeling (TRPA1 nucleofection: p = 0.0004, Mann–Whitney test; n = 50 cells per genotype, N = 5 independent cultures each; AnxA2/Mock conucleofection: p = 0.0115, Mann–Whitney test; n = 40 cells per genotype, N = 4 independent cultures each). Scale bar, 10 μm. All data are represented as mean ± SEM. C, Representative whole-cell current traces of MO-gated currents at −70 mV in DRG neurons nucleofected with mTRPA1 (black trace: WT; gray trace: AnxA2^−/−). The upper bar indicates the addition of 5 μm MO to the recording chamber. D, Left, Average ± SEM of current density after MO application measured at the current peak in each genotype (p = 0.007, Student's t test; n > 12 neurons; N = 3 cultures each). Right, Time constants of MO-induced activation and inactivation measured by a mono-exponential fit to the currents obtained at −70 mV for each genotype (n.s.; n > 12 neurons; N = 3 cultures each).

Figure 7.

Enhanced TRPA1-dependent nocifensive behaviors in AnxA2^−/− mice. A, Quantification of the latency of withdrawal of both hindpaws when WT mice and AnxA2^−/− littermates were subjected to radiant heat (Heat) or punctuate mechanical pressure (Heat: n = 10 mice each; Mechanical: n = 9 mice each; n.s., Student's t test). B, Quantification of the response duration of acute nocifensive behavior over 10 min after injection of different concentrations of MO (10 mm: WT 11.8 ± 1.9 s compared with AnxA2^−/− 18.5 ± 1.6 s, n = 10 mice each; p = 0.0126, Student's t test; 30 mm: WT 59.9 ± 15.9 s, n = 8 mice compared with AnxA2^−/− 177.4 ± 28.5 s, n = 11 mice; p = 0.0009, Student's t test; 60 mm: WT compared with AnxA2^−/−, n = 7 mice each; p = 0.5064, n.s., Student's t test), and after injection of 3 μg Cap (WT compared with AnxA2^−/−; n = 9 mice each; p = 0.4040, n.s., Student's t test). C, D, WT mice and AnxA2^−/− littermates were unilaterally injected with CFA to elicit inflammatory pain. C, Quantification of the latency of nocifensive/escaping behaviors when mice were placed on a cold plate 24 h after vehicle injection (V; n = 8 mice each; n.s.) and CFA-injection (WT CFA: 165.7 ± 17.1 s, n = 15 mice; AnxA2^−/− CFA: 75.2 ± 19.3 s, n = 13 mice; **p < 0.01 comparing values of CFA of the injected paw between genotypes; ^#p < 0.05 and ^###p < 0.001 comparing values of V and CFA of the injected paw within each genotype, ANOVA with Bonferroni′s multiple-comparison test). D, Quantification of the latency of withdrawal of the injected hindpaw and noninjected hindpaw when mice were subjected to radiant heat (left) or punctuate mechanical pressure (right) after vehicle injection (n = 8 mice each; n.s.) and CFA-injection (Heat: n = 5 mice each; n.s. between genotypes; Mechanical: n = 9 mice each, n.s. between genotypes; ^####p < 0.0001 comparing values of V and CFA of the injected paw within each genotype, ANOVA with Bonferroni′s multiple-comparison test). Noninjected paws did not develop hypersensitivity after CFA and did not differ between genotypes. All data are represented as mean ± SEM. Cap, capsaicin.